US 2515730 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
T L E F N R O l COFFEE EXTRCTION PROCESS Filed Oct. 24, 1947 Tlmwmw 3% Patented July 18, 1950 COFFEE EXTRACTION PBGCESS John Ornfelt, Morristown, N. J., assignor, by mesne assignments, to American Home Foods, Inc., New York, N. Y., a corporation of New Jersey Application October 24, 1947, Serial No. 781,787
This invention relates to a coffee extraction process adapted to produce a highly concentrated extract of good flavor and aroma, which may be sold as such or may be further processed to produce a dry powdered coee extract. Socalled instant coffee have been a popular article of commerce for many years in spite of the fact that they have had less satisfactory flavor and aroma than the best freshly brewed coiee. Continuous efforts have been made in the industry to improve the flavor and aroma of such products and simultaneously to increase the yield of coee extract from the coffee bean.
Such extracts have in general been made by extracting ground roasted coffee with hot Water and concentrating or drying the resultant extract. Spray drying has been the preferred drying method because of its rapidity, but various vacuum drying methods have also been applied. Alternatively, the liquid extract has been concentrated, bottled, and sold in this form.
Itis known that by raising the temperature of the extracting water-e. g. to 160 C. or abovea high yield of soluble matter can be obtained from a given Weight of ground roasted coiee. The eect of high temperature extraction is to cause a hydrolysis of coiee material and to convert a considerable fraction of it from an insoluble to a soluble form. Treatment of the coffee, however, with water at excessively high temperatures introduces into the final yextract products having an unpleasant, acrd, empyreumatic flavor and aroma. I have found that a moderate amount of solubilization at temperatures which are not excessive not only improves the yield of extract but actually improves its avor. Whereas extraction in the usual way with water at 100 C. or slightly below will remove normally not more than percent of the weight of the roasted coee as soluble extract, and in the case of the best grades of coffee with most eilicient extraction not more than per cent, extractionV by water under superatmospheric pressure at 160 C. or higher can be made to remove as much as 50 or 55 percent of the weight of the ground roasted coffee as soluble extract.
The coffee manufacturer is thus faced with the problem of securing as high a yield as possible from his starting material without producing a product of undesirable flavor and aroma.
8 Claims. (Cl. 99- 71) tion on contact with the air (a dierent reaction from the high-temperature hydrolysis referred to above). By extracting the freshly roasted coffee, however. with a liquid conce extract containing a relatively high concentration of coffee solubles-e. g. 35 percent or more--hydrolysis of the aromaand taste-producing substances is retarded. The production of such highly concentrated infusions involves relatively extensive contact of the aqueous extraction medium with the coffee, which in itself involves the possibility of A the destruction or deterioration of the davor and aroma unless certain precautions are carefully observed.
I have discovered that by a combination of operating steps, disclosed below, it is possible to obtain a 28-40 percent yield by Weight (based on the weight of fresh ground roasted codec treated) of coffee solubles of desirable aroma and taste in the form of a highly concentrated/aqueous extract which may be sold as such, may be further concentrated for sale as a liquid.` extract, or may be dried (as by spray drying) to produce a solid soluble coffee extract. In the latter case the high concentration of the liquid infusion produced by my process has a double advantage: it greatly reduces the cost of the drying operation because `of its relatively small water content and also favors the production of a dry product of good avor and aroma because of the short duration of the drying step resulting from its low water content.
According to my process, I extract ground roasted coffee in a plurality of stages; in the initial stages I use hot water under pressure in the temperature range of 125-150 C.; in the intermediate stages I allow the infusion to cool down progressively to a temperature preferably of approximatelv 40 C.; and iinally I use the resulting eiiiuent to extract coffee in the temperature range of approximately 25-15 C. The hot water in passing through these extraction stages is made to contact progressively less exhausted coffee; on its introduction into the extraction train it is brought into contact with nearly exhausted coffee, and prior to its exit from the train, i. e. in the above noted low temperature range,`it is contacted with unextracted ground roasted codec. Throughout the operation the extraction is effected at all stages in the substantial absence of air, and the extract on leaving the train is preferably bottled at once or run directly to a drier` to avoid as far as possible contact with' the air in the presence of a substantial amount of water.
In my process the infusion becomes progressively more concentrated as it moves through the extraction train. I have found it desirable to use a rate of flow such that on emergence from the high temperature section of the extraction it has .a concentration in the range 8-12 percent of solids by weight, and just prior to entering the final percolators at the cold end of the extraction train it has a concentration of at least 35 percent by weight of solids, so that a final extract is obtained having a minimum concentration of 40 percent of soluble coffee solids and preferablyof 50 percent or more.
In order to further illustrate my process, a preferred method of operation will be described in connection with the attached drawing which illustrates in conventional diagrammatic form one embodiment of apparatus which I may use. The description and drawing are intended to be exemplary only, and not to limit the scope of my invention, which is defined in the attached claims.
In the drawing, the percolators are indicated at Ia, Ib, Ic Ip, Iq, Ir Iy, Iz, the initial percolator Ia being shown partly in section. Each percolator is provided at its bottom with a bottom screen 2a and a top screen 2b of sufiicient fineness to prevent coffee from passing through but being readily permeable to water. A charge of ground roasted coffee 3 is supported on screen 2a. It is desirable, but not essential, that the con'ee nearly ill the percolator. The coffee may be introduced from a hopper 4 through manhole 5. The hopper may be movably mounted so that it can be brought successively over any desired percolator. Each percolator is constructed with a bottom outlet of any well-known design so that the exhausted coffee may be discharged into a suitable chute 6 and any desired disposition thereoi' made. Cold water is fed to the system from water-main 1, and is pumped to the extraction train by variable-delivery pump 8. In the hot section of the extraction train a series of heat exchangers 9a, 9b, 9c is provided in series with the percolators. as shown. Each heat exchanger is provided with a live steam inlet I and outlet II. Extraction water passes through suitable piping I2 from the variable delivery pump through heat exchanger 9a, where it is heated to a temperature between 125 and 150 C., preferably approximately 150, and at this temperature passes through pipe I 3 into the bottom of the initial percolator Ia.. The percolators are insulated, but some drop in temperature occurs as the water passes through them. This drop is made good by the supply of additional heat to the weak infusion passing from the top of the initial percolator through pipe Il by passage through the second heat exchanger 9b before introduction into th bottom of the succeeding percolator I b.
In practical operation I have found it desirable to conduct the extraction in a series of 12 to 16 percolators, in which case the hot section desirably consists of 4-7 percolators with associated heat exchangers. 0n leaving the hot section the infusion passes through pipe I into the bottom of the first percolator In of the cooling section, upwards through the coffee in that percolator, out of the top of the percolator, and so, successively, through the percolators in this section, which may advantageouslv number 5-8. No heat is introduced into the circulating infusion in this section. and the temperature is allowed to fall gradually below 50 C., e. g. to approximately 40 C. On leaving the last percolator in the cooling section by pipe I6, the infusion is 4 passed through a heat exchanger I1 provided with cold water inlet I8 and outlet I9, in which it is cooled substantially below room temperaturas. g. to approximately 15 C. It is led from thisfheat exchanger through pipe I9 into the rst percolator of the cold section, thence through another heat exchanger in which it is cooled again to approximately 15 C., and thence to the final percolator, from which it is discharged for further prompt processing as desired through pipe vAZI). The cold section advantageously comprises 2 4 percolators.
For the sake of clarity, only' suicient piping and only a suflicient number of heat exchangers have been shown in the drawing to carry out one complete cycle of extraction. In actual practice, however, and for the purposes of continuous operation, a heat exchanger having both steam and cold-water connections is provided Vfor each percolator and such piping connections and valves areprovided that any heat exchanger may be bypassed. 'Connections and valves are also provided so that 'any one of the percolators may serve as the initial percolator and any one as the final percolator, and so that intermediate percolators may be fed with infusion in the desired order and may be operated either hot or cold or with spontaneous cooling, as desired. Since such piping systems form no part of this invention and are well known in the chemical engineering art, further detailed description and illustration are believed to be unnecessary.
In starting up a system such as that above described, the percolators must, of course, be initially charged with fresh ground roasted coffee. After a, steady state of continuous operation has been reached, however, the following procedure is followed: When the coffee in the initial percolator Ia has become exhausted, circulation of infusion is halted, this percolator is cut out of the circulating system by suitable manipulation of valves, and its contents is dumped. An additional percolator, previously charged with fresh ground roasted coffee from hopper 4 through manhole 5, is connected to the exit end of the system'and circulation of infusion is resumed, cold concentrated infusion being thus introduced gradually into the bottom of the freshly charged percolator. As the liquid rises in the percolator, air is vented through pipe 2I until it has all been displaced. Carbon dioxide gas is evolved by the coffee as the particles become wetted, and this gas forms a, cushion between the upper surface of the rising infusion in the percolator and the air above it. Analysis has shown that when a percolator has been about two-thirds filled with infusion the gas above the liquid contains 80 to 85 percent CO2. When the air has been expelled from the percolator, the valve in vent 2l is closed and extraction proceeds in the normal way, the freshly charged percolator becoming the final unit in the extraction train. It is fed with highly concentrated infusion at a temperature between 25 and 15 C., and delivers the finished product. 'I'he original initial percolator Ia having been removed from the system and emptied, the original second percolator Ib becomes the new initial percolator, and so on progressively through the system, the circulating water picking up more and more `solubles from the coffee and coming progressively into contact with less and less exhausted coffee.
As an example oi' my operations I have obtained excellent results in operating such a system consisting of 14 percolators 6 in the hot section, 6 in the intermediate cooling section. and 2 in' the cold section. I charge each percolator in turn as above described with 1000 pounds of fresh ground roasted coffee, introduce water into the initial percolator at approximately 150 C., and maintain it at or near this temperature during its passage through the hot section. I regulate the rate of input flow at approximately 100 gallons per hour, and produce an infusion as it leaves the hot section containing 8-12 percent soluble solids. On its exit from the intermediate cooling section, the infusion will have picked up sufficient additional soluble solid material to bring its concentration up to nearly 35 percent or more as it enters the cold section, and on leaving the cold section it will contain from 40 to 50 percent by weight of solid soluble coffee material. The extraction cycle lasts 3 to 4 hours, at the end of which period an exhausted percolator is shunted out and a freshly charged percolator shunted into the train. I have found that in operating this way and not appreciably exceeding a temperature of 150 C. I obtain a yield of 32 to 36 percent by weight of the soluble solids contained in the coffee subjected to extraction without introducing unacceptable acrid empyreumatic flavorand aroma-producing material into the extract. I attribute this largelyto the fact that the high concentration of the extract used to extract fresh coffee at the cold end of my extraction train tends to inhibit hydrolysis of the desirable material, to the fact that such a concentrated extract is a better solvent for the desirable aromaand flavor-producing materials than is a weak infusion, to the rigid exclusion of air throughout my extraction process, and to the low temperature at which the ilnal extraction is effected and at which the final infusion leaves the extraction train. Y
By regulating the input of Vhot water into an extraction train such as that described at a rate oi 100 gal/hr. and observing the disclosed temperature limitations. I have been able to obtain extracts of excellent flavor and aroma having a soluble concentration of approximately 40-50 percent by weight and containing in the neighborhood of a 40 percent yield based on the weight of ground roasted coffee treated, i. e. each 1000 pounds of coffee will yield approximately 900 pounds of concentrate containing approximately 400 pounds of soluble solids and 500 pounds (60 gallons) of water. The specific gravity of such an infusion is subject to some variation with different grades of coii'ee and different degrees of roasts, but usually lies in the range 20o 1.21 t 1.22 -n C.
for a 45 percent concentration. 'Ihe volume of 900 pounds oi' such an infusion is thus approximately 90 gallons.
Of the 350 odd gallons of water introduced into the extraction system during a 3% hour cycle, the balance not recovered as infusion at the cold end is discharged from the system with the exhausted grounds. Since this liquid contains only a fraction of a percent of solubles of low taste value, it may be discarded.
As will be clear from the preceding description to those skilled in the extraction art, numerous changes may be made in my method and the associated apparatus without departing from my invention. For example the size, shape and number of percolators used and the rate of ilow of infusion may be varied, depending on output desired, area and height available for the equipment, and similar practical considerations. My process is, however, not a simple extraction, but a combination of extraction and controlled conversion of insoluble coffee material to soluble material by hydrolysis at a regulated high temperature: it also involvesuse of conditions which ensure maintenance of taste and aroma in the end product. It is essential, therefore, to operate within the temperature ranges indicated and to provide a degree of contact under those conditions which will result in the production of an infusion containing at least 40 percent by weight of soluble coffee extract. It is highly desirable for this purpose that contact of coffee and infusion should be so regulated as to attain interf mediate infusion concentrations, i. e. at the outlets of the hot section and the intermediate cooling section, within the limits'indicated.
If the process is so regulated as to produce too high an extraction at the hot end of the system, higher yields may be obtained but the quality of the final product will suffer. An overall yield of coffee solubles representing 40 percent of the weight offthe fresh ground roasted coeetreated is the maximum I have been able to obtain without undue sacrifice of quality; some of the advantages of my process may be realized with yields as low as 28 percent. Associated with this range of yields, extract concentrations from 40 to 50 and even 55 percent by weight of coffee solubles are obtainable and desirable.
l. In a coiee extraction process in which ground roasted coffee ds extracted in a series of stages -with water at an elevated temperature in the range of 125-150 C. to produce a highly concentrated coffee extract, the further step of using said extract in the approximate temperature range 25-15 C to extract unextracted ground roasted coil'ee.
2. In a coffee extraction process in which a highly concentrated coffee infusion of soluble unoxidized partially heat-hydrolyzed coee solids has been produced by extracting ground roasted coii'ee in a series Ofstages with water in the temperature range 125-150 C., the steps which comprise: passing said infusion into contact with unextracted ground roasted coffee at a temperature within the approximate range 25- 15 C. until lit contains in solution at least 40 percent by weight of coffee solids, air being substantially excluded from access to the infusion and to the co'ee during said contact, separating the resulting infusion from the partially extracted coffee, and further extracting the partially extracted coffee w-ith water in the range l25150 C.
3. In a process for preparing a concentrated aqueous infusion of ground roasted coffee adapted to be used for the preparation of a concentrated soluble coffee product, in which process hot water is passed serially through a plurality of percolators containing ground roasted coffee in a series of progressively less exhausted stages of extraction in the direction of travel of the water, the steps which comprise: initially extracting partially exhausted coffee with water at l25150 C., using resultant efiiuent to extract less exhausted coffee at descending temperatures in the range 12525, the volume of water and time of extraction being so regulated as to produce an eiliuent containing at least 35 percent by weight of coffee solids, and nally 7 using said resultant .eniuent at approximately 25 C. and containing at least 35 percent by weight of conee solids to extract unextracted conee, the contact of liquid with conee in all stages of the extraction being enected in the substantial absence of air.
4. In a process for preparing a concentrated aqueous infusion of ground roasted conee adapted to be used for the preparationv of a concentrated soluble conee product, in which process hot water is passed serially through a plurality of percolators containing ground roasted conee in a series of processiveiy less exhausted stages of extraction in the direction of travel of the water, the s-teps which comprise: passing water at 125-150 C. through a closed percolator containing nearly exhausted conee to produce a weak conee extract, passing the resulting extract serially through further closed percolators con-l taining less exhausted conee while maintaining the temperature of the extract in the range 125-150 C. by passage through heat exchangers between passages through percolators, thereby building up the strength of the extract, passing the resulting extract serially through additional closed percolators containing partially extracted conee while permitting the extract to cool to a temperature substantially about C., the volume of water and time of extraction being so regulated as to produce an eilluent containing at least percent by weight of conee solids, then cooling said eilluent containing at least 35 percent by weight of conee solids to a temperature within the range 25-15 C. and passing it through a closed percolator containing unextracted ground roasted conce, the contact of conee with liquid in all stages of the extraction being enected in the substantial absence of air and in later stages of the extraction under a substantial partial pressure of carbon dioxide.
5. The process as defined in claim 4, in which the rate of flow of water is so regulated as to produce a final extract containing at least percent by weight of soluble conee solids.
6. 'I'he process as denned in claim 4, in which fresh conce is intermittently added at the outlet end and exhausted conee intermittently removed from the inlet end of the extraction train and the over-all amount of conee and volume of water used are so regulated as to produce an extract containing at least 40 percent by weight oi' soluble coffee solids representing between 28 and `40 percent by weight of the conee extracted.
7. In a. process for preparing a concentrated aqueous infusion of ground roasted conee adapted to be used for the preparation of a concentrated soluble conee product, in which process hot water is passed serially through a plurality of percolators containing ground roasted conee in a series of progressively less exhausted stages o! extraction in the direction of travel of the water, the steps which comprise: passing liquid water in the temperature range of 125-150 C. into contact with ground roasted conce in a series of extraction stages in which the conee is progressively less exhausted in the direction of now of the water until the water has formed an infusion containing in solution 8 to 12 percent by weight of conee solids, passing the resultant infusion into contact with conce in a further series of extraction stages in which the conee is progressively less 'exhausted in the direction of ow of the infusion until the infusion contains in solution at least 35 percent by weight of conee solids while permitting the temperature of the infusion to fall below C. but not as low as 25 C., cooling the resulting infusion to a temperature in the range 25-15 C., and passing it into contact with unextracted ground roasted conee until it contains in solution at least 40 percent by weight of conee solids, air being substantially excluded from contact with the infusion and with the conce at all stages of the extraction.
8. In a process for preparing a cencentrated aqueous infusion of ground roasted coffee adapted to be used for the preparation of a concentrated soluble conce product, in which process hot water is passed serially through a. plurality of percolators containing ground roasted conee in a series of progressively less exhausted stages of extraction in the direction of travel of the water, the steps which comprise: extracting partially exhausted ground roasted conce with liquid water in the temperature range l25150 C. until the water has acquired a soluble solids content in the range 8-12 percent by weight, extracting partially but less exhausted ground roasted conce with the resulting infusion at temperatures in the range below that of the previous extraction but about 25 C. until the infusion has acquired a soluble solids content of at least 35 percent by weight, extracting unextracted conee with resulting infusion at a temperature in the range 25-15 C. -until it has acquired a soluble solids content of at least 40 percent by weight, and throughout the operation excluding air from contact with the infusion. and with the ground roasted conce undergoing extraction.
REFERENCES CITED The following references are of record in the le of this patent:
UNITED STATES PATENTS Number Name Date 617,322 Duke Jan. 10, 1899 1,393,045 Scott Oct. 11, 1921 1,687,112 Slocum et al. Oct. 9, 1928 1,891,383 Ginen et al Dec. 20, 1932 2,324,526 Morgenthaler July 20, 1943 2,333,027 Morgenthaler Oct. 26, 1943